1. Field of the Invention
The present invention relates to a process for producing dimethylnaphthalene (hereinafter sometimes abbreviated to "DMN"). DMN is a compound of utmost importance as a starting raw material for naphthalene dicarboxylic acid to be used in the production of plastics such as polyesters. For example, polyethylene 2,6-naphthalene dicarboxylate which is produced from 2,6-naphthalene dicarboxylic acid and ethylene glycol has heat resistance and mechanical properties more favorable than those of polyethylene terephthalate and is used for producing films and fibers.
2. Description of the Related Art
Isomerically high purity is required for naphthalene dicarboxylic acid and DMN as starting raw materials for plastics. Specifically DMN has 10 isomers according to the positions of the methyl groups, including 1,2-, 1,3-, 1,4-, 1,5-, 1,6-, 1,7-, 1,8-, 2,3-, 2,6- and 2,7-DMN and, when used as the starting raw material for naphthalene dicarboxylic acid, it is required to be a specific DMN free from other position isomers.
As the process for producing DMN, there are available an isolation process from a high boiling fraction in a petroleum refinery or the tar fraction of coal origin, an alkylation process of naphthalene, a synthetic process using an alkylbenzene and an olefin and the like.
In the case of the isolation process from a high boiling fraction in a petroleum refinery or the tar fraction of coal origin, each of the fractions is a mixture of various DMN isomers and therefore, is required to be isomerized and isolated from the resultant isomers for the purpose of obtaining the specific desired DMN from the isomers mixture. With regard to isomerization, it is well-known that the above-mentioned 10 DMN isomers are classified into 4 groups including A to D groups as mentioned hereinbelow and that the isomerization in the same group is relatively easy, whereas among different groups isomerization is difficult. In addition, it is extremely difficult to isolate the specific desired DMN from various DMN isomers. Furthermore, a variety of components other than DMN contained in the above-mentioned fractions makes it extremely difficult to isolate and recover the specific desired DMN in high purity from the mixture of DMN and the others.
Group A - - - 1,5-DMN; 1,6-DMN; and 2,6-DMN
Group B - - - 1,7-DMN; 1,8-DMN; and 2,7-DMN
Group C - - - 1,3-DMN; 2,3-DMN; and 1,4-DMN
Group D - - - 1,2-DMN
The alkylation process of naphthalene is put into practice usually by using a solid acid as the catalyst such as zeolite and silica-alumina. The process, however, involves the problems in that there are produced monomethylnaphthalene, trimethylnaphthalene, etc. other than DMN, a high selectivity to DMN is not attained and the resultant DMN is a mixture of a number of isomers. Accordingly, the process makes it difficult to afford the specific desired DMN in high yield as is the case with the isolation process from a high boiling fraction in a petroleum refinery or the tar fraction of coal origin.
As the countermeasure against the aforementioned problems, there is available a process for producing a specific DMN from an alkylbenzene and an olefin through multistage steps, exemplified by Japanese Patent Application Laid-Open No. 96540/1990 in which 2,6-DMN is produced from m-xylene, propylene and carbon monoxide and U.S. Pat. No. 5,008,479 in which 2,6-DMN is produced from toluene, butene and carbon monoxide.
Similarly, Japanese Patent Application Laid-Open Nos. 134634/1974, 89353/1975 and 67261/1974 disclose a process for producing 5-(o-tolyl)-penta-2-ene from o-xylene and butadiene, a process for producing 1,5-dimethyltetralin by cyclizing 5-(o-tolyl)-penta-2-ene and a process for producing 1,5-DMN by dehydrogenating 1,5-dimethyltetralin, respectively. The combination of the above-disclosed processes enables the production of 1,5-DMN with isomerically high purity from o-xylene and butadiene.
Japanese Patent Application Laid-Open No. 503389/1989 discloses a process for producing highly pure 2,6-DMN by isomerizing 1,5-DMN into the mixture of 1,5-DMN, 1,6-DMN and 2,6-DMN, which mixture is crystallized into the objective 2,6-DMN. The aforesaid process is highly advantageous in that isomerization and crystallization are carried out among 3 DMN isomers belonging to the same group as compared with those among the isomers belonging to different groups.
2,6-DMN has attracted the highest attention recently among the DMN isomers as the starting raw material for 2,6-naphthalene dicarboxylic acid. Thus the development of a process for industrially producing 2,6-DMN is eagerly desired. The aforesaid Japanese Patent Application Laid-Open No. 134634/1974 also discloses a process for producing 5-(p-totyl)-penta-2-ene from p-xylene and butadine. In this case, it is presumed that 1,7-DMN is obtained by the successive cyclization and dehydrogenation, 2,7-DMN is obtained in high purity by further isomerization and crystallization, and also the use of m-xylene enables the production of the mixture of 1,6-DMN and 1,8-DMN.
It can be said that the process for producing DMN by the use of xylene and butadiene as starting raw materials is industrially excellent, since it enables the production of a specific DMN with isomerically high purity as described hereinbefore.
The process for producing DMN from xylene and butadiene comprises the steps of synthesizing 5-tolyl-penta-2-ene by side-chain alkenylation, synthesizing dimethyltetralin by means of cyclization, synthesizing DMN by means of dehydrogenation, isomerizing DMN and crystallizing isolation. The synthesis of dimethyltetralin by cyclizing 5-tolyl-penta-2-ene is disclosed in Japanese Patent Application Laid-Open No. 93348/1974 in which is used a solid phosphoric acid as the catalyst and Published International Patent Application No. 500052/1991 in which is employed as the catalyst a ultra-stabilized Y-type zeolite that is modified with platinum and copper, showing a yield of 95% or more in the working examples of both the disclosures. The synthesis of DMN by dehydrogenating dimethyltetralin is disclosed in Japanese Patent Application Laid-Open Nos. 76852/1973 and 67261/1973 in which are used as the catalysts chromia-alumina and palladium with rhenium supported on a carrier, respectively, showing a yield of 95% or more in the wording examples of both the disclosures.
In addition to the high yield attained in both the cyclization and dehydrogenation steps, an attempt is made to simultaneously effect cyclization and dehydrogenation reactions, exemplified by Japanese Patent Application Laid Open No. 31151/1975 in which DMN is produced by cyclizing dehydrogenizing 5-(o-tolyl)-penta-2-ene by the use of a chromina-supporting silica-alumina based catalyst at a conversion efficiency of 97% and a DMN yield of 75% as the working example. There are also disclosed in Japanese Patent Application Laid-Open Nos. 1036/1975, 1037/1975 and 17983/1975, processes for producing DMN by cyclizing dehydrogenating 5-(o-tolyl)-penta-2-ene by the use of palladium supported on alumina or activated carbon, the combination of palladium, rhodium and rhenium supported on silica-alumina, and platinum supported on silica-alumina, respectively as the catalyst. However, in each of the working examples of the aforementioned disclosures, DMN yield is only 60% at the maximum. As is seen from the above, the two-stage process in which cyclization and dehydrogenation reactions are carried out separately attains an overall yield of 90% or more through both the reactions; while the single-stage process in which cyclization and dehydrogenation reactions are effected simultaneously attains a yield of 75% at the most. In view of the reaction performance, the two-stage process is superior to the single-stage process. Nevertheless, if a single-stage process is materialized which attains a yield comparable to that by a two-stage process, it is greatly advantageous from the industrial viewpoint in that the production process can be simplified.
In addition, the cyclization reaction is an exothermic reaction accompanied by a great reaction heat of 20 kcal/mol, causing a serious problem in the process of how to remove the reaction heat. On the contrary, the dehydrogenation reaction is an endothermic reaction accompanied with a great heat absorption of about 30 kcal/mol, bringing about a serious problem in the process of how to supply the required heat, fundamentally different from the cyclization reaction. Accordingly, the materialization of the effective single-stage process capable of simultaneously proceeding with both the reactions, if possible, greatly favors the effective utilization of reaction heat and heat engineering in addition to the process simplification.
For example, in the process for producing a specific DMN from xylene and butadiene in a multistage steps, the conventional technique of preparing DMN from the corresponding 5-tolyl-penta-2-ene formed by side-chains alkenylation requires two-stage reactions in view of the reaction performance including the cyclization of 5-tolyl-penta-2-ene into dimethyltralin, followed by dehydrogenation thereof into the corresponding DMN. If the cyclization dehydrogenation step incorporating both the reactions can be realized, it enables process simplification, effective utilization of reaction heat and rationalization of heat balance, thereby enhancing the industrial significance of itself. Thus, there is desired the development of an effective process for converting 5-tolyl-penta-2-ene into the corresponding DMN by single-stage cyclization dehydrogenation reactions.
The catalyst for converting 5-tolyl-penta-2-ene into the corresponding DMN in single stage synthesis needs to be endowed with both cyclizing and dehydrogenating functions. It is possible to cyclize 5-tolyl-penta-2-ene by the use of a solid catalyst such as solid phosphoric acid, zeolite and silica-alumina. However, a simple combination of the above-mentioned solid catalyst with the dehydrogenation catalyst typified by platinum, rhenium, palladium, rhodium, chromia-alumina, etc. can not constitute an effective cyclization dehydrogenation catalyst. It is clear from the fact that a high cyclization dehydrogenation yield is not achieved in Japanese Patent Publication No. 31151/1975 wherein chromia is supported on silica-alumina catalyst and in Japanese Patent Publication Nos. 1036/1975, 1037/1975 and 17983/1975 wherein the catalysts each comprising palladium rhodium, rhenium or platinum supported or alumina, silica-alumina or activated carbon are employed.
Moreover, only a low cyclization dehydrogenation yield is obtained even by the use of the catalyst comprising the component having a high dehydrogenating function such as palladium, rhenium, rhodium or platinum supported on the solid acid catalyst that is effective for cyclization reaction such as X-type zeolite, ultra-stabilized Y-type zeolite (hereinafter referred to as "USY"), Y-type zeolite or alumina in cyclizing 5-tolyl-penta-2-ene. It is presumed to be due to the fact that the conventional supporting method cannot increase the noble-metal supporting quantity of the resultant catalyst, thereby failing to enhance the dehydrogenating function of the catalyst. Although it is possible to enhance the dehydrogenating function to some extent by raising the reaction temperature in the aforesaid method, the elevated temperature causes decomposition and polymerization during the reaction, resulting in failure to improve the cyclization dehydrogenation yield. In other words, the catalyst of the conventional supporting system is devoid of the balance among various catalytic functions of the prepared catalyst including cyclizing activity, dehydrogenating activity, decomposing activity and polymerizing activity, thus making itself unsuitable for the objective cyclization dehydrogenation of 5-tolyl-penta-2-ene. The failure to attain a high cyclizing dehydrogenating function by the simple combination of a cyclization catalyst with a dehydrogenation catalyst is attributable also to the great difference between the reaction conditions effective for cyclization regarding reaction pressure, reaction temperature and contact time and the reaction conditions effective for dehydrogenation.